The telecommunications industry is in the process of developing a new generation of flexible and affordable communications that includes high-speed access while also supporting broadband services. Many features of the third generation mobile telecommunications system have already been established, but many other features have yet to be perfected.
One of the systems within the third generation of mobile communications is the Universal Mobile Telecommunications System (UMTS) which delivers voice, data, multimedia, and wideband information to stationary as well as mobile customers. UMTS is designed to accommodate increased system capacity and data capability. Efficient use of the electromagnetic spectrum is vital in UMTS. It is known that spectrum efficiency can be attained using frequency division duplex (FDD) or using time division duplex (TDD) schemes. Space division duplex (SDD) is a third duplex transmission method used for wireless telecommunications.
As can be seen in FIG. 1, the UMTS architecture consists of user equipment 102 (UE), the UMTS Terrestrial Radio Access Network 104 (UTRAN), and the Core Network 126 (CN). The air interface between the UTRAN and the UE is called Uu, and the interface between the UTRAN and the Core Network is called Iu.
The UTRAN consists of a set of Radio Network Subsystems 128 (RNS), each of which has geographic coverage of a number of cells 110 (C), as can be seen in FIG. 1. The interface between the subsystems is called lur.
Each Radio Network Subsystem 128 (RNS) includes a Radio Network Controller 112 (RNC) and at least one Node B 114, each Node B having geographic coverage of at least one cell 110. As can be seen from FIG. 1, the interface between an RNC 112 and a Node B 114 is called Iub, and the Iub is hard-wired rather than being an air interface. For any Node B 114 there is only one RNC 112. A Node B 114 is responsible for radio transmission and reception to and from the UE 102 (Node B antennas can typically be seen atop towers or preferably at less visible locations). The RNC 112 has overall control of the logical resources of each Node B 114 within the RNS 128, and the RNC 112 is also responsible for handover decisions which entail switching a call from one cell to another or between radio channels in the same cell.
LTE, or Long Term Evolution (also known as 3.9G), refers to research and development involving the Third Generation Partnership Project (3GPP) aimed at identifying technologies and capabilities that can improve systems such as the UMTS. LTE, or Long Term Evolution (also known as 3.9G), refers to research and development involving the Third Generation Partnership Project (3GPP) aimed at identifying technologies and capabilities that can improve systems such as the UMTS. The present invention is related to LTE work that is taking place in 3GPP (see Appendix A).
Generally speaking, a prefix of the letter “E” in upper or lower case signifies LTE. The E-UTRAN consists of eNBs (E-UTRAN or enhanced Node Bs), providing the E-UTRA user plane (RLC/MAC/PHY) and control plane (RRC) protocol terminations towards the UE. The eNBs interface to the access gateway (aGW) via the S1, and are inter-connected via the X2.
An example of the E-UTRAN architecture is illustrated in FIG. 1b. This example of E-UTRAN consists of eNBs, providing the E-UTRA user plane (RLC/MAC/PHY) and control plane (RRC) protocol terminations towards the UE. The eNBs are interconnected with each other by means of the X2 interface. The eNBs are also connected by means of the S1 interface to the EPC (evolved packet core), e.g. to the MME (mobility management entity) and the UPE (user plane entity) which may form the access gateway (aGW). The S1 interface supports a many-to-many relation between MMEs/UPEs and eNBs. The S1 interface supports a functional split between the MME and the UPE.
In this example there exists an X2 interface between the eNBs that need to communicate with each other. For exceptional cases (e.g. inter-PLMN handover), LTE_ACTIVE inter-eNB mobility is supported by means of MME/UPE relocation via the S1 interface.
The eNB may host functions such as radio resource management (radio bearer control, radio admission control, connection mobility control, dynamic allocation of resources to UEs in both uplink and downlink), selection of a mobility management entity (MME) at UE attachment, routing of user plane data towards the user plane entity (UPE), scheduling and transmission of paging messages (originated from the MME), scheduling and transmission of broadcast information (originated from the MME or O&M), and measurement and measurement reporting configuration for mobility and scheduling. The MME/UPE may host functions such as the following: distribution of paging messages to the eNBs, security control, IP header compression and encryption of user data streams; termination of U-plane packets for paging reasons; switching of U-plane for support of UE mobility, idle state mobility control, SAE bearer control, and ciphering and integrity protection of NAS signaling.
A particular aspect of LTE is control channel structure for cellular systems. In such a system, there are multiple resources for which an individual channel quality indicator can be reported. A typical example would be the allocation of frequency domain resources LTE. Let us assume that there are multiple resources that can be allocated to a user, and that the user can provide channel quality indication (CQI) for these multiple resources. A scheduler then has the ability to assign the channels to the users based on this resource-specific CQI.
In LTE, the present working assumption is that a set of 12 subcarriers or 180 kilohertz (kHz) is the bandwidth of a Physical Resource Block (PRB), which is the minimum resource that can be allocated to a user when a localized allocation principle is used, “localized” meaning that full PRBs are allocated to users. In a 10 megahertz (MHz) bandwidth, there would be fifty (50) such PRBs.
It is important to understand that PRBs can be allocated via unrestricted allocation, or via adjacent allocation. The extremes of allocation freedom are as follows. For unrestricted allocation, a user can get any number of disjoint resource blocks; so, user 1 could get, for example, resource blocks 1, 2, 3, 6,10,11,15,16,18, and 22. For adjacent allocation, a user can get any number of adjacent resource blocks, but no disjoint sets. So, user 1 could get, for example, consecutive resource blocks 4, 5, 6, 7, 8, 9, 10, and in this case his allocation could be extended with PRB 3 and/or PRB 11, but not, for example, with non-consecutive PRB 44. Unrestricted allocation is the working assumption in LTE downlink (DL), whereas adjacent allocation is the working assumption in LTE uplink (UL). In channels with a high degree of frequency selectivity, it is understood that an unrestricted allocation gives up to 50% higher throughput than an adjacent allocation principle.
In 3GPP radio access network discussions (RAN1), two options for coding of the physical layer (L1) and data link layer (L2) allocation-related control information have been discussed: separate coding and joint coding. According to separate coding, the allocation entries for individual users are separate packets, and the user is not assumed to need to know anything about another user's allocations in order to be able to process his own allocation. In contrast, according to joint coding, all (or at least some) users are coded together, so that the user needs some information about the other users, such as the number of other users jointly coded, and the other user's ordinal number in the set, in order to be able to identify his own allocation. In joint coding, the number of information bits is smaller than in separate coding. In separate coding, however, it is easier to adapt the power usage to the channel condition of the user.
It is preferable to have fixed-length signaling fields. For unrestricted allocation with separate coding, the simplest (and only) fixed-length signaling method is that a bit mask is transmitted for each user so that “0” indicates that the user does not get the resource, whereas “1” indicates that the user gets the resource. For instance, in the LTE 10 MHz option, 50 bits per user equipment (UE) would be needed in order to signal an unrestricted allocation with separate coding.
In joint coding, the simplest signaling method uses a UE index. First there is a (typically implicit) mapping, where each UE is assigned a number. Thus UE1 would have the index 0, UE2 the index 1, and so on. For each of the 50 resources, the index of the UE which gets the allocation is indicated. Various methods to indicate that a resource is not allocated at all have been devised. For example, the index 0 can be used to indicate “no allocation”. Alternatively, a separate bit field per resource can be used to indicate that a resource is or is not allocated. To gain more from joint coding, the UE indexes may be reported in a non-binary alphabet. For example, if one out of three alternatives is chosen, then one may use a three-valued object to select the alternative for each PRB.
For further background information related to the invention, incorporated by reference herein in its entirety is the document TSG-RAN WG1 #44, R1-060573 (Feb. 13-17, 2006) titled “E-UTRA Downlink Control Signaling—Overhead Assessment”, which discusses various aspects of joint versus separate coding. Also incorporated by reference herein in its entirety is the document 3GPP TSG-RAN WG1 LTE Ad Hoc, R1-061907 (Jun. 27-30, 2006) titled “DL L1/L2 control signaling channel encoding structures”, which discusses various joint coding structures.
The problem addressed by the present invention is to reduce the signaling overhead related to the unrestricted allocation principle, without losing the flexibility and gains as compared to the adjacent allocation principle. This problem is especially relevant in the case of separate coding, where a bit field of 50 bits per UE starts to be in the upper limit of tolerability.
In the past, there have been attempts to solve this problem. For background about those past attempts, incorporated by reference herein in its entirety is the document TSG-RAN WG1 LTE Ad Hoc Meeting, R1-060094 (Jan. 23-25, 2006) titled “Signaling of E-UTRA Scheduling Information”, which discusses various aspects of joint versus separate coding. In that document, it was proposed for there to be a run length and index for each UE signaled. Run length Ru means that for UE u, only sets of R consecutive resources are allocated. There is no restriction that all such sets should be adjacent. Thus if Ru is one for all users, the method is the same as the “unrestricted method”, with the added overhead of signaling Ru. For example, the run length for UE 1 is 4, and for UE 2 the run length is 3. The indexes for UE1 and UE2 would be 0 and 1, respectively. In the allocation table, there is a list of UE indexes: 0 0 1 0 1 0 0 0. Each “0” means 4 consecutive resources allocated to UE1, and each “1” means 3 consecutive resources allocated to UE 2. Thus after expanding according to the run length, the resource allocation would be as follows:
0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0. So, altogether, 24 resources are allocated to UE1, and 6 resources to UE2.
This type of prior art does not apply to separate coding at all. From a joint coding perspective, it is difficult to implement at the transmitter. If one wants to gain from the restrictions, one needs to solve a constrained optimization problem, where the run length of all users affects the resources available for all other users. A further difficulty with that proposed solution is that the run length of each user also needs to be signaled to all users in order for each UE to know how to decode the allocation table.
Background about a second attempt to solve this problem of reducing signaling overhead can be found in the document 3GPP TSG-RAN WG1 LTE Ad Hoc, R1-061801 (Jun. 27-30, 2006) titled “Multiplexing of L1/L2 Control in E-UTRA DL” which is also incorporated herein by reference in its entirety. According to this second attempted solution, one is restricting the possible control signals to a high degree. By allowing only a set of distributed type of allocations, 9 bits are used to signal an allocation in 5 MHz (12 resource blocks). Those 9 bits include 3 bits allocated sub-band map, plus 2 bits starting RB ID in the first sub-band, plus 2 bits ending RB ID in the last sub-band, plus 2 bits spacing of allocated RBs (“spacing” indicates a code in the frequency domain). This second type of prior art allows only very limited gains over “adjacent” allocation. Essentially, this second type of prior art method selects an adjacent set of resources for the users, and then divides the resources in the code domain, so that the user does not take all the resources in his adjacent band. General background about technology related to the present invention can be found in the U.S. Application of Tsuyoshi Kashima and Sigit Jarot, titled “Control Signal Structure for Resource Allocation In E-UTRA” (Ser. No. 11/486,834), which is incorporated herein by reference in its entirety.